RadGenome-Anatomy: A Large-Scale Anatomy-Labeled Chest Radiograph Dataset via Physically Grounded Volumetric Projection

TL;DR

This paper introduces RadGenome-Anatomy, the largest annotated chest radiograph dataset created by projecting 3D anatomical masks from CT scans into 2D radiographic space, enabling improved medical image analysis.

Contribution

It presents a novel large-scale dataset with physically grounded annotations, shifting from 2D tracing to volumetric projection for better anatomical labeling.

Findings

Over 10 million segmentation masks across 210 structures

Achieved high diagnostic accuracy for cardiomegaly, kyphosis, and scoliosis

Enables research on geometric measurements for radiograph interpretation

Abstract

Anatomical structure labels for chest radiographs are essential for medical image segmentation and a broad range of downstream diagnostic tasks. However, annotating anatomy directly on 2D chest radiographs is labor-intensive and intrinsically ambiguous, as 3D anatomical structures are projected onto a single 2D plane where boundaries may overlap, be occluded, or appear only partially visible. Consequently, existing anatomy-labeled chest radiograph datasets remain limited in scale, anatomy coverage, and label reliability. To address these limitations, we introduce RadGenome-Anatomy, the largest anatomy-labeled chest radiograph dataset, containing over 10 million segmentation masks across 210 anatomical structures in 25,692 studies. It is constructed by projecting large-scale 3D anatomical masks from CT volumes into 2D radiographic space through canonical radiographic geometry. This shifts annotation from directly tracing uncertain 2D boundaries to defining anatomy in volumetric space, where structures that overlap or become partially invisible in radiographs remain spatially separable. As a result, each 2D mask represents the physically grounded projected footprint of a volumetrically defined structure. The scale and broad anatomical coverage of RadGenome-Anatomy, including structures that are overlapping, partially visible, or difficult to delineate directly, enable research on geometric measurements as explicit evidence for chest radiograph interpretation. We demonstrate this by training XAnatomy to predict structure-specific masks and derive clinically relevant measurements, achieving diagnostic accuracies of 96.4%, 95.6%, and 89.2% for cardiomegaly, kyphosis, and scoliosis, respectively.